Development of a vaccine against the filoviruses Ebola (EBOV) and Marburg (MARV) is hindered by the simultaneous circulation of multiple species and strains of these viruses, including those with no or limited antigenic relatedness. Vaccine candidates currently in development include the GP protein as the major protective antigen of filoviruses. Each of these candidates has severe limitations. These include the need for very high or multiple doses for protection, the lack of immunogenicity in the presence of vector-specific antibodies, the potential virulence of vectors, high cost of production, and the lack of feasible technologies for their needle-free administration. In addition, the most important limitations of the existing vaccine candidates is that they all require multiple components to protect against individual species of filoviruses, and their unclear level of protection against currently unidentified species or strains which will appear in the future and have a limited antigenic relatedness to the known species. Our previous data demonstrate that mucosal respiratory tract immunization of guinea pigs and non-human primates (NHP) with vectors based on attenuated respiratory paramyxoviruses expressing EBOV GP induces a robust serum EBOV-neutralizing antibody response; confers to the animals sterilizing immunity against intraperitoneal challenge with a highly lethal dose of EBOV; and induces mucosal antibody response in the respiratory tract. Recent breakthrough approaches for development of single-component vaccines against the H5N1 highly pathogenic avian influenza viruses were based on the generation of consensus, ancestral or Computationally-Optimized Broadly Reactive Antigen (COBRA) sequences derived from multiple viruses, which conferred protection against multiple diverse clades of the virus with no or limited antigenic relatedness. Moreover, a past study with influenza virus demonstrated that a mucosal, but not parenteral delivery of an influenza vaccine induced cross-reactive mucosal IgG and IgA and serum IgG, and B-cell dependent protection against antigenically different heterosubtypic influenza viruses. The proposal includes development of a broadly specific, pan-filovirus vaccine based on ancestral, consensus, or COBRA GP proteins of EBOV and MARV, expressed by attenuated respiratory paramyxoviruses, whose breadth of protection will be further enhanced by their mucosal administration through the respiratory tract, and which will be protective against all filoviruses that circulate at the present time as well as those that will circulate in the future. The Specific Aims are the following: (1) Develop and test single vaccine components capable of inducing multivalent responses against EBOV or MARV by deriving ancestral, consensus, or COBRA GP proteins for each of the two viruses; (2) Improve immunogenicity and protective efficacy of the vaccine, and minimize time required for induction of the protective immune response by selecting the most potent vector variants and optimizing dosage and the vaccination regimen; (3) Analyze the systemic and local immune responses to immunizations to determine valid correlates of protection for clinical trials.